Abstract
Subsurface fibrous capillary irrigation has a great potential for high water-saving capability compared to other irrigation methods. However, it suffers from under-performance due to inaccurate estimation and manipulation of water supply depth (∆h) in the water wetting zone of the plant root, in accordance with the changing dynamics of soil, plant growth, and weather. This work presents the design and experimental implementation of Kalman filter-proportional integral derivative (KF-PID) controller for the subsurface fibrous capillary irrigation system. This controller is designed to control the optimal water supply depth (∆h) between the fibrous capillary interface and the surface of the water, through which water is wicked through the fibrous capillary interface to the root zone of the plant. The designed Kalman filter (KF) uses a dynamic model of the system which was obtained using historical real-time data through data-driven system identification. This yields a state-space model to describe the plant's dynamic characteristics. In addition, the KF will reduce the noise sensor output which will improve the accuracy of ∆h estimation. Furthermore, a PID controller was designed to control minimizes the error between the estimated ∆h and reference ∆h to find an optimal value for ∆h. An experimental investigation on the performance comparison of the designed controller on the cultivation of the Mustard leaf plant is also presented in this paper. To access the performance of the designed control system, three performance indices are used to compare the performance of the proposed controller with a fuzzy controller through simulation using integral absolute error (IAE), integral square error (ISE), and integral absolute square error (ITAE). The results show that the proposed controller has lower values of these indices compared to the fuzzy logic controller. Also, the proposed KF-PID controller was able to estimate the optimal ∆h accurately to ensure supply of water to the fibrous capillary material for effective wetting of the plant root zone. This is evident from water saved in treatment A controlled by KF-PID which is 56.3% greater than the treatment B controlled by the adaptive fuzzy logic. The KF-PID controller shows a slightly better performance in terms of controller performance and water productivity index (WPI) which is 16 g/ Liters higher than the adaptive fuzzy logic controller.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have